Abstract
Objective
To examine the association between gestational age at birth and hospital admission costs from birth to 8 years of age.
Design
Population-based, record linkage, cohort study in England.
Setting
National Health Service (NHS) hospitals in England, UK.
Participants
1 018 136 live, singleton births in NHS hospitals in England between 1 January 2005 and 31 December 2006.
Main outcome measures
Hospital admission costs from birth to age 8 years, estimated by gestational age at birth (<28, 28–29, 30–31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41 and 42 weeks).
Results
Both birth admission and subsequent admission hospital costs decreased with increasing gestational age at birth. Differences in hospital admission costs between gestational age groups diminished with increasing age, particularly after the first 2 years following birth. Children born extremely preterm (<28 weeks) and very preterm (28–31 weeks) still had higher average hospital admission costs (£699 (95% CI £419 to £919) for <28 weeks; £434 (95% CI £305 to £563) for 28–31 weeks) during the eighth year of life compared with children born at 40 weeks (£109, 95% CI £104 to £114). Children born extremely preterm had the highest 8-year cumulative hospital admission costs per child (£80 559 (95% CI £79 238 to £82 019)), a large proportion of which was incurred during the first year after birth (£71 997 (95% CI £70 866 to £73 097)).
Conclusions
The association between gestational age at birth and hospital admission costs persists into mid-childhood. The study results provide a useful costing resource for future economic evaluations focusing on preventive and treatment strategies for babies born preterm.
Keywords: Health services research, Neonatology, Child Health, Health Care Economics and Organizations
WHAT IS ALREADY KNOWN ON THIS TOPIC.
Most previous research on the economic consequences associated with gestational age focused on the costs of initial birth admission or costs incurred during the first few years after birth.
A small number of studies have examined the association between gestational age at birth and hospital costs over the longer term, but were based on regional data, decision-analytic models synthesising summary evidence from multiple sources, cross-sectional assessments at specific ages, or focused on narrow categories of the gestational age range.
WHAT THIS STUDY ADDS
Using a large national cohort with hospital records linked from birth until mid-childhood, our study quantifies hospital admission costs from birth up to 8 years of age across the full range of gestational age at birth.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
The results should act as a useful resource for clinical and budgetary service planning, and as data inputs for economic evaluations of preventive and treatment strategies for babies born at different gestational ages.
Background
The rates of preterm birth (<37 weeks’ gestation) have increased or remained stable over the past few decades in most countries,1–3 accounting for 10.6% of all live births worldwide in 2014.3 Survival rates following preterm birth have increased as a result of technological advances,4 5 but these babies still remain at a higher risk of infant mortality and a range of short-term and long-term morbidities.6 7 A recent study examined the association between gestational age at birth and hospital admissions and found that gestational age at birth is a strong predictor of severe morbidity throughout childhood, even for those born at 38 and 39 weeks’ gestation.8
Most previous research on the economic consequences associated with gestational age focused on the costs of initial birth admission or costs incurred during the first few years after birth.9 A small number of studies have investigated hospital costs by gestational age over the longer term, but were based on historical region-specific data,10 decision-analytic models synthesising summary evidence from multiple sources,11 12 cross-sectional assessments at a specific age,13 or focused on narrow categories of the gestational age spectrum.14 To the best of our knowledge, no study has comprehensively estimated the long-term economic burden associated with gestational age from birth to middle childhood across the full spectrum of gestational age using a large national cohort of children born in the 21st century.
In this study, we conducted an evidence synthesis exercise to examine the association between gestational age at birth and hospital admission costs from birth to 8 years of age using a population-based, record-linkage study that included all live, singleton births occurring in England in 2005/2006 as part of the TIGAR study (Tracking the Impact of Gestational Age on Health, Educational and Economic outcomes: a Longitudinal Records Linkage Study).8
Methods
Data sources
In this study, we synthesised data from three data sources: TIGAR, National Neonatal Research Database (NNRD) and Paediatric Intensive Care Audit Network (PICANet) databases. Access to individual patient-level data was available for the TIGAR dataset, whereas aggregate data were available from the NNRD and PICANet datasets.15 16
The TIGAR cohort was built through a population-based data linkage using data from the Office for National Statistics (ONS) birth registration records linked to death registration records, birth notification records and Hospital Episode Statistics Admitted Patient Care (HES APC) records.17 A description of the datasets, linkage and quality assurance has been published elsewhere.8 18 19 In brief, the TIGAR cohort included all live, singleton births occurring in an NHS hospital in England between 1 January 2005 and 31 December 2006 with follow-up capturing all inpatient admissions to National Health Service (NHS) hospitals in England from birth until 31 March 2015. Children were not eligible if they had opted out, died before discharge from the birth admission or if there were data quality issues.8
The NNRD is a national resource holding quality-assured real-world clinical data captured during the course of care for all admissions to NHS neonatal units in England, Wales, Scotland and the Isle of Man.15 PICANet is an audit database recording demographic and clinical information on all patients admitted to paediatric intensive care units in the UK and Ireland.16
Study design
The main source of information for our economic analysis was the TIGAR dataset. However, although in the HES APC records the length of stay for admissions includes any time spent in critical care units, the information did not reliably indicate the level of care the child received on a day-to-day basis (online supplemental material 1.1). Therefore, we requested bespoke aggregate tables from the NNRD and PICANet to supplement the individual-level data within the TIGAR cohort. This information was used together with the total number of live births in England by gestational age in the same year, as reported by ONS, to simulate the following information for children in the TIGAR cohort by sex and gestational age at birth at an individual level20:
fetalneonatal-2022-324763supp001.pdf (1.2MB, pdf)
Whether a child was admitted to a neonatal or paediatric critical care unit during the birth admission.
The number of days that were spent in a neonatal or paediatric critical care unit during the birth admission by level of care.
Any difference between the total birth admission days observed in the TIGAR cohort and the critical care days estimated from the NNRD and PICANet was considered non-critical care ward days. More details about the methods used to simulate neonatal and paediatric critical care days and calculate non-critical care ward days in birth admission can be found in online supplemental materials 1.2 and 1.3.
Costs
We estimated direct costs from a healthcare perspective in England. In the HES APC, each data record indicates a Finished Consultant Episode, which represents a continuous period of care under one clinical consultant. Costs were estimated at episode level. The 2018–2019 Casemix Grouper Software (HRG4+) was used to allocate each episode to a Healthcare Resource Group (HRG), primarily based on any procedures performed, diagnoses, hospital admission type, episode length of stay and patient characteristics.21 HRGs are standard groupings of clinically similar treatments, which use comparable levels of healthcare resource. The NHS 2017–2018 reference cost schedules were used to price the HRGs.22 More details on the application of the Grouper Software and matching the HRGs to reference costs can be found in online supplemental materials 2 and 3.
Neonatal and paediatric critical care costs during the birth admission were calculated on a per diem basis using the NHS 2017–2018 reference costs based on level of care. Birth admission ward costs were adjusted accordingly by extracting days spent in critical care units from the total hospital stay (online supplemental material 1.3).
Statistical analysis
We used descriptive statistics to estimate the average total length of stay and average costs for the birth admission and for subsequent hospital admissions by year of follow-up among children alive at the beginning and not censored by the end of each follow-up year. Comparisons were made between the following groups by gestational age at birth (weeks): <28, 28–29, 30–31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42. Gestational ages were grouped <28, 28–29 and 30–31 following ONS policy about reporting of small numbers.
We used the Kaplan-Meier sample average estimator to calculate the total 1-year, 5-year and 8-year costs for each gestational age group.23 The 95% CIs for total costs and p values for cost differences between gestational age at birth using 40 weeks as the reference were obtained using non-parametric bootstrapping with 1000 replications. No adjustment based on baseline characteristics was conducted.
In the baseline analysis, we calculated the birth admission non-critical care ward days and costs according to aggregated data from the NNRD and PICANet. We conducted a sensitivity analysis to investigate the results when allocating all the estimated non-critical care ward days to neonatal critical care for children born at ≤33 weeks to reflect an scenario where these babies spent most of their stay in a neonatal unit as previously indicated.24
All analyses were conducted using Stata V.14 (College Station, Texas, USA). Costs are presented in 2018 UK pounds (£).
Results
A total of 1 018 136 children were included in this study, with a total of 9 372 105 person years of follow-up and 1 315 338 admissions that occurred between 1 January 2005 and 31 March 2015. The baseline characteristics of the TIGAR cohort are presented in table 1. Among the 1 018 136 children included in this study, 56 053 (5.5%) were born at <37 weeks, and 99 717 (9.8%) were small for gestational age at birth (birth weight below the 10th centile).
Table 1.
Baseline characteristics of live born singletons (total N=.1 018 136)
| n | % | |
| Mother’s age at childbirth (years) | ||
| <20 | 44 486 | 4.4 |
| 20–24 | 181 633 | 17.8 |
| 25–29 | 253 055 | 24.9 |
| 30–34 | 293 741 | 28.9 |
| 35–39 | 193 622 | 19.0 |
| 40+ | 51 599 | 5.1 |
| Parity | ||
| Nulliparous | 480 616 | 47.2 |
| Parous | 496 203 | 48.7 |
| Missing | 41 317 | 4.1 |
| Mother’s marital status | ||
| Married | 581 160 | 57.1 |
| Partner | 347 366 | 34.1 |
| Single | 89 610 | 8.8 |
| Mother’s country of birth | ||
| Non-UK | 225 695 | 22.2 |
| UK | 791 012 | 77.7 |
| Missing | 1429 | 0.1 |
| Index of Multiple Deprivation* | ||
| Least deprived Q1 | 276 838 | 27.2 |
| Q2 | 216 006 | 21.2 |
| Q3 | 180 300 | 17.7 |
| Q4 | 161 793 | 15.9 |
| Most deprived Q5 | 157 195 | 15.4 |
| Missing | 26 004 | 2.6 |
| Sex | ||
| Male | 521 169 | 51.2 |
| Female | 496 967 | 48.8 |
| Ethnicity (child) | ||
| White British | 677 236 | 66.5 |
| White other | 59 683 | 5.9 |
| Bangladeshi | 14 546 | 1.4 |
| Indian | 27 783 | 2.7 |
| Pakistani | 41 739 | 4.1 |
| Black African | 34 571 | 3.4 |
| Black Caribbean | 12 410 | 1.2 |
| Other | 91 570 | 9.0 |
| Missing | 58 598 | 5.8 |
| Gestational age (weeks) | ||
| <28 | 1730 | 0.2 |
| 28–29 | 2089 | 0.2 |
| 30–31 | 3227 | 0.3 |
| 32 | 2656 | 0.3 |
| 33 | 4050 | 0.4 |
| 34 | 7292 | 0.7 |
| 35 | 11 663 | 1.1 |
| 36 | 23 346 | 2.3 |
| 37 | 54 001 | 5.3 |
| 38 | 137 926 | 13.5 |
| 39 | 231 376 | 22.7 |
| 40 | 288 065 | 28.3 |
| 41 | 208 757 | 20.5 |
| 42 | 41 958 | 4.1 |
| Small for gestational age at birth | ||
| No | 918 419 | 90.2 |
| Yes | 99 717 | 9.8 |
| Mode of birth | ||
| Vaginal | 751 653 | 73.8 |
| Caesarean section | 222 615 | 21.9 |
| Missing | 43 868 | 4.3 |
| Labour induction | ||
| No | 626 178 | 61.5 |
| Yes | 154 851 | 15.2 |
| Missing | 237 107 | 23.3 |
| Season of birth | ||
| Jan–Mar | 236 944 | 23.3 |
| Apr–Jun | 254 016 | 24.9 |
| Jul–Sep | 270 282 | 26.5 |
| Oct–Dec | 256 894 | 25.2 |
*The Index of Multiple Deprivation (IMD) is the official measure of relative deprivation for small areas (or neighbourhoods) in England.
The proportions of children admitted to neonatal or paediatric critical care during the birth admission are presented in table 2. The admission rates decreased with increasing gestational age at birth, with the lowest rate observed at 39–40 weeks (table 2). Children born extremely preterm (<28 weeks) spent, on average, 50.2 (SD 18.8) days in neonatal critical care, which generated an average cost of £53 144 (SD 15 504); in comparison, 8% of children born at 40 weeks were admitted to neonatal critical care with an average stay of 4.0 (SD 2.5) days, which generated an average cost of £2369 (SD £1783). Further, 5.5% of children born extremely preterm and 0.1% of children born at 40 weeks were admitted to paediatric critical care during their birth admission, which generated an average cost of £14 967 (SD £9526) and £21 957 (SD £14 523), respectively. The average hospital days and costs incurred as a result of admission to non-critical care wards also decreased with increasing gestational age at birth (table 3), with the longest stay and highest cost observed for the extremely preterm group (48.8 days, £13 142) and the shortest stay and lowest cost observed for those born at 40 gestational weeks (1.3 days, £244). The total costs associated with birth admissions including critical and non-critical care is presented in table 4.
Table 2.
Proportion of children admitted to neonatal critical care (NCC) and paediatric critical care (PCC) during the birth admission and, among those admitted, length of stay and cost, by gestational age at birth
| % to NCC | Days in NCC, mean (SD) | Costs in NCC, mean (SD) | % to PCC | Days in PCC, mean (SD) |
Costs in PCC, mean (SD) | |
| <28 | 100% | 50.2 (18.8) | £53 144 (15 504) | 5.5% | 8.8 (5.9) | £14 967 (9526) |
| 28–29 | 100% | 39.3 (13.8) | £32 717 (10 185) | 1.7% | 9.6 (4.0) | £18 265 (9166) |
| 30–31 | 100% | 30.2 (9.3) | £21 957 (6753) | 1.7% | 9.0 (4.0) | £17 596 (8361) |
| 32 | 100% | 22.0 (6.2) | £15 350 (4772) | 0.4% | 16.1 (6.5) | £28 530 (10 386) |
| 33 | 100% | 16.3 (4.4) | £11 065 (3516) | 0.4% | 12.5 (7.8) | £24 168 (15 387) |
| 34 | 90% | 11.9 (4.4) | £7883 (3277) | 0.4% | 13.3 (7.0) | £24 390 (13 400) |
| 35 | 63% | 7.7 (3.9) | £5018 (2981) | 0.4% | 13.5 (6.6) | £24 087 (13 943) |
| 36 | 40% | 5.7 (3.4) | £3757 (2658) | 0.4% | 11.2 (6.8) | £20 069 (13 466) |
| 37 | 20% | 5.0 (2.9) | £3343 (2419) | 0.2% | 9.4 (5.0) | £19 003 (11 033) |
| 38 | 11% | 4.3 (2.8) | £2821 (2302) | 0.2% | 9.5 (4.9) | £19 225 (11 420) |
| 39 | 8% | 4.0 (2.5) | £2417 (1798) | 0.1% | 10.6 (5.9) | £23 720 (15 050) |
| 40 | 8% | 4.0 (2.5) | £2369 (1783) | 0.1% | 9.8 (5.9) | £21 957 (14 523) |
| 41 | 10% | 4.0 (2.4) | £2411 (1721) | 0.1% | 5.3 (3.7) | £13 080 (12 786) |
| 42 | 12% | 4.1 (2.6) | £2383 (1825) | 0.1% | 5.5 (3.2) | £10 936 (9374) |
Table 3.
Non-critical ward days and costs during the birth admission, by gestational age at birth
| n | Mean days in non-critical care ward | Mean costs in non-critical care ward | 95% CI | |
| <28 | 1730 | 48.8 | £13 142 | £12 512 to £13 771 |
| 28–29 | 2089 | 23.4 | £5567 | £5310 to £5823 |
| 30–31 | 3227 | 9.7 | £2394 | £2291 to £2498 |
| 32 | 2656 | 5.4 | £1377 | £1311 to £1443 |
| 33 | 4050 | 3.7 | £1016 | £963 to £1068 |
| 34 | 7292 | 1.8 | £532 | £509 to £556 |
| 35 | 11 663 | 2.4 | £809 | £775 to £843 |
| 36 | 23 346 | 2.2 | £753 | £725 to £781 |
| 37 | 54 001 | 1.8 | £463 | £446 to £480 |
| 38 | 137 926 | 1.8 | £373 | £365 to £382 |
| 39 | 231 376 | 1.6 | £285 | £279 to £290 |
| 40 | 288 065 | 1.3 | £244 | £240 to £249 |
| 41 | 208 757 | 1.4 | £262 | £258 to £267 |
| 42 | 41 958 | 1.5 | £342 | £329 to £356 |
Table 4.
Birth admission costs and costs in subsequent admissions by year of follow-up and gestational age
| Birth admission | Year 1 | Year 2 | Year 3 | Year 4 | Year 5 | Year 6 | Year 7 | Year 8 | ||
| <28 | n | 1730 | 1730 | 1710 | 1705 | 1701 | 1697 | 1695 | 1695 | 1691 |
| Mean | £67 168 | £4885 | £2726 | £1539 | £1347 | £933 | £848 | £656 | £669 | |
| SD | £33 071 | £10 937 | £11 342 | £6446 | £8032 | £4895 | £7961 | £4333 | £5248 | |
| 28–29 | n | 2089 | 2089 | 2071 | 2069 | 2067 | 2067 | 2066 | 2065 | 2064 |
| Mean | £38 616 | £2893 | £1410 | £716 | £703 | £567 | £587 | £564 | £522 | |
| SD | £18 893 | £7610 | £6564 | £3033 | £4419 | £4252 | £4289 | £5511 | £4719 | |
| 30–31 | n | 3227 | 3227 | 3218 | 3217 | 3214 | 3213 | 3212 | 3211 | 3211 |
| Mean | £24 640 | £2029 | £754 | £636 | £421 | £527 | £434 | £330 | £377 | |
| SD | £12 231 | £5698 | £3309 | £5178 | £2225 | £4082 | £4119 | £2142 | £4805 | |
| 32 | n | 2656 | 2656 | 2646 | 2645 | 2644 | 2642 | 2642 | 2641 | 2639 |
| Mean | £16 803 | £1813 | £842 | £592 | £539 | £442 | £370 | £344 | £240 | |
| SD | £8056 | £6004 | £4578 | £3955 | £4104 | £2837 | £2899 | £2094 | £1267 | |
| 33 | n | 4050 | 4050 | 4026 | 4021 | 4018 | 4017 | 4016 | 4015 | 4015 |
| Mean | £12 212 | £1387 | £667 | £399 | £311 | £325 | £256 | £232 | £194 | |
| SD | £7316 | £5931 | £4234 | £2819 | £1695 | £2733 | £1571 | £1458 | £1218 | |
| 34 | n | 7292 | 7292 | 7255 | 7247 | 7244 | 7242 | 7240 | 7240 | 7240 |
| Mean | £7732 | £1399 | £655 | £396 | £338 | £251 | £278 | £262 | £209 | |
| SD | £6570 | £7269 | £5340 | £3547 | £3224 | £1829 | £2982 | £2316 | £1755 | |
| 35 | n | 11 663 | 11 663 | 11 614 | 11 601 | 11 593 | 11 590 | 11 588 | 11 588 | 11 583 |
| Mean | £4052 | £1086 | £503 | £311 | £275 | £259 | £243 | £226 | £218 | |
| SD | £6990 | £5883 | £4195 | £2625 | £2636 | £2759 | £2801 | £2622 | £2701 | |
| 36 | n | 23 346 | 23 346 | 23 273 | 23 255 | 23 249 | 23 246 | 23 239 | 23 237 | 23 235 |
| Mean | £2325 | £1030 | £437 | £314 | £264 | £279 | £242 | £205 | £181 | |
| SD | £6235 | £6061 | £3141 | £3468 | £2378 | £3700 | £2913 | £2901 | £2454 | |
| 37 | n | 54 001 | 54 001 | 53 889 | 53 865 | 53 845 | 53 836 | 53 824 | 53 818 | 53 812 |
| Mean | £1166 | £744 | £356 | £247 | £233 | £201 | £196 | £167 | £159 | |
| SD | £4709 | £4012 | £2953 | £2296 | £2516 | £1921 | £2482 | £1677 | £1887 | |
| 38 | n | 137 926 | 137 926 | 137 711 | 137 654 | 137 616 | 137 596 | 137 575 | 137 563 | 137 553 |
| Mean | £725 | £570 | £291 | £208 | £191 | £178 | £164 | £147 | £143 | |
| SD | £3839 | £3438 | £2347 | £2079 | £1990 | £1765 | £1642 | £1608 | £1952 | |
| 39 | n | 231 376 | 231 376 | 231 150 | 231 072 | 231 028 | 230 993 | 230 970 | 230 950 | 230 932 |
| Mean | £502 | £453 | £239 | £175 | £160 | £160 | £148 | £127 | £121 | |
| SD | £3058 | £2962 | £2157 | £1800 | £1569 | £1703 | £1743 | £1484 | £1612 | |
| 40 | n | 288 065 | 288 065 | 287 821 | 287 748 | 287 691 | 287 663 | 287 644 | 287 627 | 287 607 |
| Mean | £455 | £394 | £227 | £164 | £149 | £142 | £133 | £122 | £109 | |
| SD | £2860 | £2690 | £2071 | £1899 | £1496 | £1412 | £1401 | £1467 | £1352 | |
| 41 | n | 208 757 | 208 757 | 208 576 | 208 528 | 208 501 | 208 480 | 208 460 | 208 447 | 208 436 |
| Mean | £519 | £358 | £219 | £158 | £151 | £154 | £135 | £120 | £115 | |
| SD | £2585 | £2419 | £1898 | £1614 | £1494 | £1926 | £1582 | £1426 | £1519 | |
| 42 | n | 41 958 | 41 958 | 41 917 | 41 902 | 41 896 | 41 890 | 41 888 | 41 881 | 41 880 |
| Mean | £641 | £391 | £224 | £159 | £168 | £152 | £131 | £118 | £116 | |
| SD | £2849 | £2935 | £1702 | £1360 | £2056 | £1582 | £1302 | £1523 | £1637 |
The TIGAR cohort excludes babies who died within their birth admission. As a result, for subsequent admissions all babies were alive at the beginning of the first year.
Full details on estimated lengths of stay for admissions subsequent to the birth admission, by year of follow-up, can be found in the online supplemental materials 4. Associated average hospital costs for these subsequent admissions are presented in table 4. Similar to birth admission costs, subsequent hospital admission costs decreased with increasing gestational age at birth, with the largest differences between gestational age groups observed during the first year after birth. Differences in hospital costs between gestational age groups diminished with increasing age, particularly during the first 2 years after birth, while children born extremely preterm (<28 weeks) and very preterm (28–31 weeks) still had higher hospital admission costs (£699 (95% CI £419 to £919) for <28 weeks; £434 (95% CI £305 to £563) for 28–31 weeks) during the eighth year of life compared with children born at 40 weeks (£109, 95% CI £104 to £114).
The mean cumulative hospital admission cost over 8 years after birth among children born extremely preterm was estimated at £80 559 (95% CI £79 238 to £82 019) per child, with most of the cost incurred during the first year after birth (£71 997 (95% CI £70 866 to £73 097)) (table 5). Children born at 40 weeks’ gestational age incurred the lowest 1-year, 5-year and 8-year cumulative hospital admission costs compared with other gestational age groups, with all cost differences being statistically significant (table 5). Even children born at 39 weeks had a higher 8-year cumulative hospital admission cost (£2085 (95% CI £2061 to £2107)) compared with those born at 40 weeks (£1894 (95% CI £1874 to £1912)). In the sensitivity analysis, after allocating all the non-critical care ward days to neonatal critical care for children born at ≤33 weeks, estimated total hospital admission costs were higher at these gestational ages, especially for the extremely preterm group where the 8-year total hospital admission cost increased to £119 044 (95% CI £117 350 to £120 909) per child.
Table 5.
Total cumulative 1-year, 5-year and 8-year hospital cost (£) by gestational age at birth, estimated with the Kaplan-Meier sample-average estimator
| 1-year total | 5-year total | 8-year total | |||||||
| Mean | 95% CI | P value | Mean | 95% CI | P value | Mean | 95% CI | P value | |
| <28 | 71 997 | 70 866 to 73 097 | <0.0001 | 78 432 | 77 164 to 79 818 | <0.0001 | 80 559 | 79 238 to 82 019 | <0.0001 |
| 28–29 | 41 484 | 40 843, 42 135 | <0.0001 | 44 846 | 44 097 to 45 584 | <0.0001 | 46 499 | 45 705, 47 291 | <0.0001 |
| 30–31 | 26 663 | 26 310 to 26 989 | <0.0001 | 28 991 | 28 551 to 29 440 | <0.0001 | 30 126 | 29 612 to 30 605 | <0.0001 |
| 32 | 18 609 | 18 313 to 18 899 | <0.0001 | 21 013 | 20 613 to 21 447 | <0.0001 | 21 961 | 21 550 to 22 409 | <0.0001 |
| 33 | 13 591 | 13 353 to 13 821 | <0.0001 | 15 280 | 14 979 to 15 570 | <0.0001 | 15 955 | 15 637 to16 257 | <0.0001 |
| 34 | 9124 | 8918 to 9310 | <0.0001 | 10 755 | 10 495 to 11 007 | <0.0001 | 11 498 | 11 214, 11 769 | <0.0001 |
| 35 | 5133 | 5001 to 5257 | <0.0001 | 6474 | 6303 to 6638 | <0.0001 | 7156 | 6967 to 7334 | <0.0001 |
| 36 | 3351 | 3262 to 3438 | <0.0001 | 4641 | 4523 to 4758 | <0.0001 | 5266 | 5131 to 5399 | <0.0001 |
| 37 | 1909 | 1866 to 1951 | <0.0001 | 2943 | 2883 to 3002 | <0.0001 | 3464 | 3402 to 3534 | <0.0001 |
| 38 | 1294 | 1273 to 1315 | <0.0001 | 2161 | 2128 to 2191 | <0.0001 | 2612 | 2576 to 2646 | <0.0001 |
| 39 | 955 | 940 to 970 | <0.0001 | 1689 | 1669 to 1710 | <0.0001 | 2085 | 2061 to 2107 | <0.0001 |
| 40 | 848 | 837 to 861 | Reference | 1530 | 1512 to 1547 | Reference | 1894 | 1874 to 1912 | Reference |
| 41 | 877 | 865 to 889 | <0.0001 | 1558 | 1540 to 1577 | 0.021 | 1928 | 1905 to 1950 | 0.013 |
| 42 | 1032 | 997 to 1066 | <0.0001 | 1732 | 1685 to 1777 | <0.0001 | 2096 | 2045 to 2148 | <0.0001 |
| Sensitivity analysis* | |||||||||
| <28 | 110 482 | 108 946 to 112 149 | <0.0001 | 116 917 | 115 267 to 118 675 | <0.0001 | 119 044 | 117 350 to 120 909 | <0.0001 |
| 28–29 | 55 365 | 54 629 to 56 171 | <0.0001 | 58 727 | 57 893 to 59 638 | <0.0001 | 60 380 | 59 443 to 61 404 | <0.0001 |
| 30–31 | 31 332 | 30 939 to 31 712 | <0.0001 | 33 660 | 33 191 to 34 136 | <0.0001 | 34 794 | 34 273 to 35 336 | <0.0001 |
| 32 | 20 982 | 20 656 to 21 303 | <0.0001 | 23 386 | 22 918 to 23 834 | <0.0001 | 24 334 | 23 838 to 24 804 | <0.0001 |
| 33 | 15 079 | 14 845 to 15 320 | <0.0001 | 16 768 | 16 451 to 17 059 | <0.0001 | 17 444 | 17 125 to 17 747 | <0.0001 |
*Allocate all non-paediatric critical care birth admission days to neonatal critical care stay for gestational age <=33 weeks.
Discussion
In this study, we have investigated hospital admission costs from birth to mid-childhood in England across the full range of gestational age at birth. We found that gestational age at birth is associated not only with birth admission hospital costs but also subsequent hospital admission costs up to age 8. The most common cause of subsequent admissions was infection.8 Children born extremely preterm were estimated to have high hospital admission costs throughout the first 8 years of life, with the majority of the costs incurred during the first year after birth.
Our cost estimates are in line with other studies of the costs of preterm birth in England. Khan et al reported similar birth admission hospital costs for 32–33 week moderately preterm born (£13 168) and 34–36 week late preterm born (£5463) children (2017–18 prices), based on a cohort in the East Midlands region of England.14 Mangham et al estimated the costs to the public sector over the first 18 years after birth using a decision-analytic model and reported neonatal care costs of £109 860 (2017–2018 price) for extremely preterm children.11 This is higher than our base case estimates (£53 144) when 2017 NNRD data are applied, but similar to the costs generated by our sensitivity analysis (£110 482) that assumed that the entire birth admission of children born at ≤33 weeks in TIGAR was spent in critical care.
This is, to our knowledge, the first study that uses a large national cohort to examine the association between gestational age at birth and hospital admission costs across the full spectrum of gestational age. The results of this study provide comprehensive estimates of hospital admission costs from birth to mid-childhood by gestational age at birth, which can act as a useful resource for clinical and budgetary service planning, and as data inputs for economic evaluations of preventive and treatment strategies for preterm birth.
The main strength of this study is the large sample size available for analysis, which provided a sufficient sample to estimate costs across the full spectrum of gestational age. The national coverage of the TIGAR cohort contrasts with the regional-based or clinic-based populations that provided a focus for previous studies.10 14 The linkage to routinely collected HES data accompanied by standardised costing approaches through HRG Groupers ensured an accurate estimation of hospital costs, as it based the estimation on more granular patient-level hospital activities and nationally recommended cost algorithms.
The main limitation of the study is that we were not able to estimate more granular neonatal and paediatric critical care costs for the TIGAR cohort as such estimation is not supported by the HES APC data; notably, critical care information was collected with insufficient quality in HES APC before 2008. A separate HES dataset covers adult critical care from 2008/2009, whereas data relating to neonatal or paediatric intensive care are collected through systems external to NHS Digital, which collects HES data.17 To address this issue, we conducted an evidence synthesis exercise and simulated cost estimates for neonatal or paediatric intensive care using aggregated data from the NNRD and PICANet. This allowed us to account for critical care costs in the analysis. In the case of PICANet, aggregate data were provided only by gestational week bands instead of a specific week. Therefore, we had to assume that the information provided for a specific band was the same across all weeks. This may have introduced some inaccuracies in our estimation of length of stay and associated costs for PCC. In addition, we were not able to obtain neonatal and paediatric critical care data over the same time coverage of the TIGAR cohort as such information does not date back to 2005–2006. This may have contributed to some of the differences between the total birth admission days observed in TIGAR and the neonatal and paediatric critical care days simulated using NNRD and PICANet data. Nevertheless, our sensitivity analysis that estimated costs using alternative assumptions about the ward stay during the birth admission of infants born either very or extremely preterm provides an upper bound for cost estimates for the birth admission.
It is also worth highlighting that the TIGAR cohort excluded children who died before discharge from their birth admission, which means that we have excluded a small group of extremely ill babies from our study. This suggests that our cost estimates should be viewed as conservative, particularly for the purposes of planning of neonatal services.
In conclusion, this study provides estimates of the association between gestational age at birth and hospital admission costs from birth to mid-childhood and disaggregates those estimates by gestational category and chronological year. The study results should act as a useful resource for future economic evaluations that focus on preventive and treatment strategies for preterm birth and inform resource allocation decisions.
Acknowledgments
We are grateful to the following organisations and individuals who were involved in this study: Medical Research Council (MRC) for funding this project (MR/M01228X/1); Office for National Statistics (ONS) for providing access to data and hosting it within the safe research setting (SRS); NHS Digital for linking and providing access to the data; National Neonatal Research Database (NNRD) for providing aggregated data on neonatal critical care admissions; Paediatric Intensive Care Audit Network (PICANet) for providing aggregated data on paediatric critical care admissions; the TIGAR patient and public involvement (PPI) group and National Childbirth Trust (NCT) representatives for their input into the study objectives and design; and the TIGAR advisory group for their input into the study design and dissemination. This work contains statistical data from ONS, which is Crown Copyright. Use of the ONS statistical data in this work does not imply endorsement of the ONS of the interpretation or analysis of the statistical data. This work uses research datasets, which might not exactly reproduce National Statistics aggregates.
Footnotes
Twitter: @xinyang_hua
Contributors: CRediT (Contributor Roles Taxonomy) author statement: Conceptualisation: OR-A, SP, MAQ, JJK. Methodology: XH, OR-A, SP. Formal analysis: XH with input from OR-A, CC and VC. Data curation: MAQ, AM. Writing—Original draft: XH with input from OR-A and SP. Writing—Review and editing: XH, SP, VC, CC, JJK, MAQ, EB, SJ, AM, OR-A. Visualisation: XH, OR-A. Supervision: OR-A. Project administration: OR-A, MAQ. Funding acquisition: MAQ, OR-A, SP, CC, JJK, EB, SJ, AM. OR-A acts as the guarantor.
Funding: The TIGAR study was funded by a research grant from the MRC (MR/M01228X/1). The funder had no role in the study design or in the collection, analysis, interpretation of data, writing of the report or decision to submit the article for publication. XH is an AUS NHMRC Emerging Leadership Fellow (grant number 2009253). SP receives support as a UK National Institute for Health Research (NIHR) Senior Investigator (NF-SI-0616–10103) and from the UK NIHR Applied Research Collaboration Oxford and Thames Valley.
Competing interests: None declared.
Provenance and peer review: Not commissioned; externally peer reviewed.
Supplemental material: This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
Data availability statement
Data may be obtained from a third party and are not publicly available. The authors do not have permission to supply data or identifiable information to third parties, including other researchers, but the team at City, University of London has permission under regulation 5 of the Health Service (control of patient information) Regulations 2002 to analyse patient identifiable data for England and Wales without consent and create a research database that could be accessed by other researchers using the SRS at the ONS. The TIGAR team has permission under regulation 5 of the Health Service (control of patient information) Regulations 2002 to analyse these data. Anyone wishing to access the linked datasets for research purposes should apply via the CAG to the Health Research Authority to access patient identifiable data without consent and then to the ONS and NHS Digital. In the first instance, enquiries about access to the data should be addressed to Alison Macfarlane.
Ethics statements
Patient consent for publication
Not applicable.
Ethics approval
Ethical approval for this study was granted by the Health Research Authority research ethics committee (South West–Frenchay; REC reference 15/SW/0294) and was provided by the Health Research Authority following Advice from the Confidentiality Advisory Group (CAG reference 15/CAG/0196). Participants gave informed consent to participate in the study before taking part.
References
- 1. Goldenberg RL, Culhane JF, Iams JD, et al. Epidemiology and causes of preterm birth. Lancet 2008;371:75–84. 10.1016/S0140-6736(08)60074-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Blencowe H, Cousens S, Oestergaard MZ, et al. National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: a systematic analysis and implications. Lancet 2012;379:2162–72. 10.1016/S0140-6736(12)60820-4 [DOI] [PubMed] [Google Scholar]
- 3. Chawanpaiboon S, Vogel JP, Moller A-B, et al. Global, regional, and national estimates of levels of preterm birth in 2014: a systematic review and modelling analysis. Lancet Glob Health 2019;7:e37–46. 10.1016/S2214-109X(18)30451-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Stoll BJ, Hansen NI, Bell EF, et al. Trends in care practices, morbidity, and mortality of extremely preterm neonates, 1993-2012. JAMA 2015;314:1039–51. 10.1001/jama.2015.10244 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Santhakumaran S, Statnikov Y, Gray D, et al. Survival of very preterm infants admitted to neonatal care in England 2008-2014: time trends and regional variation. Arch Dis Child Fetal Neonatal Ed 2018;103:F208–15. 10.1136/archdischild-2017-312748 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Saigal S, Doyle LW. An overview of mortality and sequelae of preterm birth from infancy to adulthood. Lancet 2008;371:261–9. 10.1016/S0140-6736(08)60136-1 [DOI] [PubMed] [Google Scholar]
- 7. Moster D, Lie RT, Markestad T. Long-Term medical and social consequences of preterm birth. N Engl J Med 2008;359:262–73. 10.1056/NEJMoa0706475 [DOI] [PubMed] [Google Scholar]
- 8. Coathup V, Boyle E, Carson C, et al. Gestational age and hospital admissions during childhood: population based, record linkage study in England (TIGAR study). BMJ 2020;371:m4075. 10.1136/bmj.m4075 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Petrou S, Yiu HH, Kwon J. Economic consequences of preterm birth: a systematic review of the recent literature (2009-2017). Arch Dis Child 2019;104:456–65. 10.1136/archdischild-2018-315778 [DOI] [PubMed] [Google Scholar]
- 10. Petrou S. The economic consequences of preterm birth during the first 10 years of life. BJOG 2005;112 Suppl 1:10–5. 10.1111/j.1471-0528.2005.00577.x [DOI] [PubMed] [Google Scholar]
- 11. Mangham LJ, Petrou S, Doyle LW, et al. The cost of preterm birth throughout childhood in England and Wales. Pediatrics 2009;123:e312–27. 10.1542/peds.2008-1827 [DOI] [PubMed] [Google Scholar]
- 12. Johnston KM, Gooch K, Korol E, et al. The economic burden of prematurity in Canada. BMC Pediatr 2014;14:93. 10.1186/1471-2431-14-93 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Petrou S, Abangma G, Johnson S, et al. Costs and health utilities associated with extremely preterm birth: evidence from the EPICure study. Value Health 2009;12:1124–34. 10.1111/j.1524-4733.2009.00580.x [DOI] [PubMed] [Google Scholar]
- 14. Khan KA, Petrou S, Dritsaki M, et al. Economic costs associated with moderate and late preterm birth: a prospective population-based study. BJOG 2015;122:1495–505. 10.1111/1471-0528.13515 [DOI] [PubMed] [Google Scholar]
- 15. Gale C, Morris I, Neonatal Data Analysis Unit (NDAU) Steering Board . The UK national neonatal research database: using neonatal data for research, quality improvement and more. Arch Dis Child Educ Pract Ed 2016;101:216–8. 10.1136/archdischild-2015-309928 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Paediatric intensive care audit network annual report (2020) (published 02/2021): universities of leeds and leicester. online. 2020.
- 17. Herbert A, Wijlaars L, Zylbersztejn A, et al. Data resource profile: Hospital episode statistics admitted patient care (Hes APC). Int J Epidemiol 2017;46:1093–1093i. 10.1093/ije/dyx015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Macfarlane A, Dattani N, Gibson R, et al. Births and their outcomes by time, day and year: a retrospective birth cohort data linkage study. Health Serv Deliv Res 2019;7:1–268. 10.3310/hsdr07180 [DOI] [PubMed] [Google Scholar]
- 19. Coathup V, Macfarlane A, Quigley M. Linkage of maternity hospital episode statistics birth records to birth registration and notification records for births in England 2005-2006: quality assurance of linkage. BMJ Open 2020;10:e037885. 10.1136/bmjopen-2020-037885 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Office for national statistics . Live births by gestational age, where mother’s usual residence was england, 2016 and 2017. ref: 009545. 2019.
- 21. National Casemix Office . Grouper user manual: HRG4+ 2018/19 reference costs grouper. 2019.
- 22. NHS . National schedule of NHS costs 2017/18. 2018.
- 23. Lin DY, Feuer EJ, Etzioni R, et al. Estimating medical costs from incomplete follow-up data. Biometrics 1997;53:419–34. [PubMed] [Google Scholar]
- 24. Seaton SE, Barker L, Draper ES, et al. Estimating neonatal length of stay for babies born very preterm. Arch Dis Child Fetal Neonatal Ed 2019;104:F182–6. 10.1136/archdischild-2017-314405 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
fetalneonatal-2022-324763supp001.pdf (1.2MB, pdf)
Data Availability Statement
Data may be obtained from a third party and are not publicly available. The authors do not have permission to supply data or identifiable information to third parties, including other researchers, but the team at City, University of London has permission under regulation 5 of the Health Service (control of patient information) Regulations 2002 to analyse patient identifiable data for England and Wales without consent and create a research database that could be accessed by other researchers using the SRS at the ONS. The TIGAR team has permission under regulation 5 of the Health Service (control of patient information) Regulations 2002 to analyse these data. Anyone wishing to access the linked datasets for research purposes should apply via the CAG to the Health Research Authority to access patient identifiable data without consent and then to the ONS and NHS Digital. In the first instance, enquiries about access to the data should be addressed to Alison Macfarlane.